← Back to Skills Marketplace
Zhipu Image Generator
by
Jiang Shuaijie
· GitHub ↗
· v1.1.1
· MIT-0
371
Downloads
1
Stars
1
Active Installs
5
Versions
Install in OpenClaw
/install zhipu-image-gen
Description
AI text-to-image generation using Zhipu AI's GLM-Image model. Use when the user requests image generation, creating images from text descriptions, or mention...
Usage Guidance
This skill appears to do exactly what it claims: call Zhipu's image API and save the returned PNG. Before installing, verify you trust the source and the API domain (open.bigmodel.cn) and avoid sharing your ZHIPU_API_KEY with untrusted parties. The script downloads whatever URL the API returns — which is normal here, but be cautious about storing your API key in shared files. If you need stronger protections, use a dedicated API key with limited quota and monitor usage in the Zhipu console.
Capability Analysis
Type: OpenClaw Skill
Name: zhipu-image-gen
Version: 1.1.1
The skill provides legitimate image generation functionality via Zhipu AI but contains a JSON injection vulnerability in `scripts/generate_image.sh`. The script constructs a JSON payload for a `curl` request by directly interpolating unescaped variables (`$PROMPT`, `$SIZE`), which could allow malformed or malicious input to manipulate the API call. Additionally, the script permits writing files to arbitrary system locations via the `-o` (output directory) parameter.
Capability Assessment
Purpose & Capability
Name/description state a Zhipu text-to-image client; required binaries (curl, jq) and required env var (ZHIPU_API_KEY) match that purpose. No unrelated credentials or tools are requested.
Instruction Scope
SKILL.md and the included script limit actions to reading a local .env (only extracting ZHIPU_API_KEY), calling the Zhipu API endpoint, parsing JSON with jq, and downloading the returned image URL. The script does not read other system files or unrelated environment variables.
Install Mechanism
Instruction-only skill with a simple shell script; there is no install step that downloads or executes arbitrary third-party code. Risk from install mechanism is minimal.
Credentials
Only a single API key (ZHIPU_API_KEY) is required and the script only uses that key for authenticating to the documented API. No other secrets, system tokens, or config paths are requested.
Persistence & Privilege
always is false and the skill does not request elevated or persistent system privileges. It does perform outbound network calls to the Zhipu API (expected for this purpose).
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install zhipu-image-gen - After installation, invoke the skill by name or use
/zhipu-image-gen - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.1.1
安全修复：1) 声明环境变量和二进制依赖元数据；2) 安全读取.env文件（仅提取ZHIPU_API_KEY）；3) 添加.env.example模板
v1.1.0
新增 watermark_enabled 参数（-w/--watermark），支持控制是否启用水印，默认关闭
v1.0.2
- Added a "安全说明" (Security Information) section detailing required permissions, their purposes, and data safety assurances.
- No other functional or usage changes.
v1.0.1
- Added detailed usage instructions and configuration steps for setting up Zhipu AI's GLM-Image text-to-image generator.
- Described command-line usage, supported prompt languages (Chinese/English), and image sizes.
- Specified dependency on ZHIPU_API_KEY, with setup instructions via `.env` file or environment variable.
- Included example prompts, output options, and notes on image format and file naming.
- Provided a clear file structure overview for easy navigation and setup.
v1.0.0
- Initial release of zhipu-image-gen skill.
- Enables AI text-to-image generation using Zhipu AI's GLM-Image model.
- Supports both Chinese and English prompts, including key phrases like "文生图", "生成图片", and "画图".
- Allows configurable image sizes and output directories.
- Simple setup via .env file or environment variable.
- Includes script and documentation for usage and configuration.
Metadata
Frequently Asked Questions
What is Zhipu Image Generator?
AI text-to-image generation using Zhipu AI's GLM-Image model. Use when the user requests image generation, creating images from text descriptions, or mention... It is an AI Agent Skill for Claude Code / OpenClaw, with 371 downloads so far.
How do I install Zhipu Image Generator?
Run "/install zhipu-image-gen" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Zhipu Image Generator free?
Yes, Zhipu Image Generator is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does Zhipu Image Generator support?
Zhipu Image Generator is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Zhipu Image Generator?
It is built and maintained by Jiang Shuaijie (@codeteenager); the current version is v1.1.1.
More Skills